U.S. health care costs may reach $2 trillion by the year 2000. Over- utilization is one of the major reasons given for rising health care costs. Illness behavior patterns, including health care utilization and disability days, run in families, and parental modeling and reinforcement have been proposed as likely mechanisms to explain family aggregation patterns. However, studies of parental reinforcement of illness behavior have been limited by retrospective reports from adults about their childhood experiences. The first aim of this study is to assess the way parents respond to their children's somatic complaints and to assess the relationship of parental reinforcement and modeling to various measures of illness behavior in their children. The long term goal is the development of an intervention model for prevention or reducing inappropriate illness behavior. A second aim is to determine whether the effects of modeling and reinforcement are independent of the psychosocial variables of family stress, competence, and parent and child psychological symptoms in determining child illness behavior. The proposed study will focus on gastrointestinal illness behavior in the children of adults with irritable bowel syndrome (IBS). Subjects will consist of children aged 8-16 years whose mothers have sought care for IBS from a defined managed care (HMO) population, control children whose mothers have sought care for asthma. All families will be randomly selected from the same population. For the IBS and non-IBS controls, questionnaire, interview and diary data will be obtained from parents and children on illness behavior, reinforcement, psychological symptoms, stress coping, and beliefs about illness, and school attendance data will be obtained from the schools. For all subjects, data will be collected from the HMO database on number, cost, type, and diagnosis for all parent and child health care visits. Data from the HMO and schools will be obtained prospectively for two years after entering the study and retrospectively for the two years prior to entering the study.